Time-Homogeneous Markov Modeling of HIV Progression in Patients Receiving Antiretroviral Therapy Treatment in the Ashanti Region, Ghana.

Abstract

The global HIV/AIDS pandemic remains a profound public health challenge, with substantial impacts on mortality and morbidity worldwide. In Ghana, where HIV prevalence persists, understanding disease progression among patients receiving antiretroviral therapy (ART) is crucial. This study, conducted in the Ashanti Region, employs a 5-state continuous-time Markov multistate model to analyze HIV progression based on CD4 cell counts, employing tuberculosis (TB) coinfection as a covariate. A retrospective cohort of 416 patients from St. Martins Catholic Hospital between 2000 and 2019 was studied. Transition intensities, sojourn time and probabilities between CD4 states, and the impact of TB coinfection were evaluated. The results showed that patients with CD4 counts ≥ 500 cells/mm³ spent more time before transitioning to lower CD4 levels, indicating the effectiveness of ART in controlling the disease at this level. However, the transition from 200-350 cells/mm³ to death was more likely than recovery to CD4 counts ≥ 500 cells/mm³, indicating the increased risk of mortality once CD4 counts drop significantly. TB coinfection did not significantly alter these transition probabilities, which may be due to the effective management of both HIV and TB in this cohort, emphasizing the need for integrated care strategies. This study emphasizes the importance of tailored interventions to manage HIV/AIDS effectively, particularly in regions with high disease burden. It is recommended that initiating treatment quickly can help maintain higher CD4 counts and improve survival.